Variational Entropic Optimal Transport

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dyachenko, Roman, Gushchin, Nikita, Sokolov, Kirill, Mokrov, Petr, Burnaev, Evgeny, Korotin, Alexander
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915768865652736
author Dyachenko, Roman
Gushchin, Nikita
Sokolov, Kirill
Mokrov, Petr
Burnaev, Evgeny
Korotin, Alexander
author_facet Dyachenko, Roman
Gushchin, Nikita
Sokolov, Kirill
Mokrov, Petr
Burnaev, Evgeny
Korotin, Alexander
contents Entropic optimal transport (EOT) in continuous spaces with quadratic cost is a classical tool for solving the domain translation problem. In practice, recent approaches optimize a weak dual EOT objective depending on a single potential, but doing so is computationally not efficient due to the intractable log-partition term. Existing methods typically resolve this obstacle in one of two ways: by significantly restricting the transport family to obtain closed-form normalization (via Gaussian-mixture parameterizations), or by using general neural parameterizations that require simulation-based training procedures. We propose Variational Entropic Optimal Transport (VarEOT), based on an exact variational reformulation of the log-partition $\log \mathbb{E}[\exp(\cdot)]$ as a tractable minimization over an auxiliary positive normalizer. This yields a differentiable learning objective optimized with stochastic gradients and avoids the necessity of MCMC simulations during the training. We provide theoretical guarantees, including finite-sample generalization bounds and approximation results under universal function approximation. Experiments on synthetic data and unpaired image-to-image translation demonstrate competitive or improved translation quality, while comparisons within the solvers that use the same weak dual EOT objective support the benefit of the proposed optimization principle.
format Preprint
id arxiv_https___arxiv_org_abs_2602_02241
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Variational Entropic Optimal Transport
Dyachenko, Roman
Gushchin, Nikita
Sokolov, Kirill
Mokrov, Petr
Burnaev, Evgeny
Korotin, Alexander
Machine Learning
Entropic optimal transport (EOT) in continuous spaces with quadratic cost is a classical tool for solving the domain translation problem. In practice, recent approaches optimize a weak dual EOT objective depending on a single potential, but doing so is computationally not efficient due to the intractable log-partition term. Existing methods typically resolve this obstacle in one of two ways: by significantly restricting the transport family to obtain closed-form normalization (via Gaussian-mixture parameterizations), or by using general neural parameterizations that require simulation-based training procedures. We propose Variational Entropic Optimal Transport (VarEOT), based on an exact variational reformulation of the log-partition $\log \mathbb{E}[\exp(\cdot)]$ as a tractable minimization over an auxiliary positive normalizer. This yields a differentiable learning objective optimized with stochastic gradients and avoids the necessity of MCMC simulations during the training. We provide theoretical guarantees, including finite-sample generalization bounds and approximation results under universal function approximation. Experiments on synthetic data and unpaired image-to-image translation demonstrate competitive or improved translation quality, while comparisons within the solvers that use the same weak dual EOT objective support the benefit of the proposed optimization principle.
title Variational Entropic Optimal Transport
topic Machine Learning
url https://arxiv.org/abs/2602.02241